Package release sniffer
Load when tracking newly published package or model-package releases across package registries and release feeds for AI/developer-tool monitoring; do not load for ordinary docs lookup, broad GitHub trend scanning, or implementing package clients.From its SKILL.md
npx -y skills add JasonxzWen/harness-hub --skill package-release-snifferAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
What its file declares
Copied from the file, not written here
The file declares its own license as MIT. That is the author’s claim about this one file, and it is not the same thing as the license GitHub reports for the repository, which is listed with the other numbers below.
SKILL.md
3.5 KB, 674 tokens by cl100k_base, as published. Nobody here has run it
Package Release Sniffer
Use this skill to find newly published packages or meaningful package releases before they become general news.
Workflow
- Define the package ecosystem and freshness window before searching.
- Check primary release surfaces first: registry package pages, registry APIs, release feeds, GitHub Releases, changelogs, model cards, and maintainer announcement posts that link to the package.
- Record the package name, ecosystem, version, publish time, source URL, maintainer or organization, license signal, install surface, and why the release matters.
- Prefer packages with concrete shipped artifacts: new package, major/minor version, security fix, SDK/API capability, model/runtime integration, benchmarked performance change, or developer workflow impact.
- De-duplicate aliases across registry pages, GitHub repositories, model cards, and maintainer posts.
- Mark uncertainty when publish time, package identity, maintainership, license, or artifact availability is unclear.
Source Priority
Use primary sources before secondary mentions:
- npm package pages and npm registry metadata
- PyPI project pages and release history
- GitHub Releases, tags, changelogs, and repository package metadata
- Hugging Face model, dataset, Space, and library release pages when they behave like package artifacts
- crates.io, Go package/module pages, Maven Central, NuGet, Docker Hub, or similar registries when relevant
- Maintainer blogs or docs only when they link back to the package or release artifact
Keep
- First public release of an AI or developer-tool package
- New major or minor release with a clear capability change
- SDK, CLI, MCP server/client, inference runtime, agent framework, eval, data pipeline, observability, RAG, vector database, or deployment package
- Security, compatibility, performance, or migration-impact release
- Package release that is not yet visible in broader trend/news sources
Drop
- Media-only claims without a package or release source
- Repository commits without a packaged artifact
- Awesome lists, prompt dumps, demo-only repos, and placeholders
- Releases with unclear maintainership, no usable artifact, or no relevant AI/developer-tool signal
- Ordinary patch releases unless they affect security, compatibility, or widely used workflows
Output Shape
Return concise candidates:
{
"package": "",
"ecosystem": "npm | pypi | github-release | hugging-face | crates | go | maven | nuget | docker | other",
"version": "",
"published_at": "YYYY-MM-DDTHH:mm:ssZ",
"source_url": "",
"maintainer": "",
"signal": "new_package | major_release | minor_release | security | performance | compatibility | ecosystem",
"summary": "",
"why_it_matters": "",
"confidence": "high | medium | low"
}
Boundaries
- Use
documentation-lookupwhen the task is to read current docs for an already chosen package. - Use
source-postwhen a confirmed release must become a publishable article. - Use implementation workflows when the user wants a registry client, scraper, monitor, or scheduled job built.
- Do not create accounts, subscribe to feeds, change automation, or publish packages.
What ships with it
Read from the repository
Just SKILL.md. No reference files, no scripts.
Gives 0 of the 12 instructions most ship operate skills give in 674 tokens
Counted across 779 of the 1,178 authors here whose files we hold, read 2026-08-07
- Document a rollback plan before deploymentin 41 of 779, across 22 files
- Update the changelogin 21 of 779, across 19 files
- Run the test suitein 20 of 779
- Create an annotated git tagin 20 of 779
- Clean up feature flags after full rolloutin 18 of 779, across 10 files
- Verify deployment health after launchin 18 of 779, across 10 files
- Test both feature flag statesin 17 of 779, across 9 files
- Verify the working tree is cleanin 17 of 779
- Make database migrations backward-compatiblein 16 of 779, across 8 files
- Set up error monitoring before launchin 15 of 779, across 7 files
- Monitor metrics at each rollout stagein 14 of 779, across 5 files
- Create a GitHub releasein 14 of 779
Said here and by no other author read
- define the package ecosystem and freshness window
- check primary release surfaces first
- record package name version time source and maintainer
- prefer packages with concrete shipped artifacts
- de-duplicate aliases across sources
- mark uncertainty when metadata is unclear
Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.